python - Fast basic linear algebra in Cython for recurrent calls -

-

i'm trying program function in cython monte-carlo simulation. function involves multiple small linear algebra operations, dot products , matrix inversions. function being called hundred of thousands of times numpy overhead getting large share of cost. 3 years ago 1 asked question: calling dot products , linear algebra operations in cython? have tried use recommendations both answers, first scipy.linalg.blas still goes through python wrapper , i'm not getting improvements. second, using gsl wrapper slow , tends freeze system when dimensions of vectors large. found ceygen package, looked promising seems installation file broke in last cython update. on other hand saw scipy working on cython wrapper lapack, looks still unavailable (scipy-cython-lapack) in end can code own c routines operations seems kind of re-inventing wheel.

so summarize: there new way kind of operations in cython? (hence don't think duplicate) or have found better way deal sort of problem haven't seen yet?

obligatory code sample: (this example, of course can still improved, give idea)

 cimport numpy np  import numpy np   cpdef double risk(np.ndarray[double, ndim=2, mode='c'] x,      np.ndarray[double, ndim=1, mode='c'] v1,       np.ndarray[double, ndim=1, mode='c'] v2):       cdef np.ndarray[double, ndim=2, mode='c'] tmp, sumx      cdef double ret       tmp = np.exp(x)      sumx = np.tile(np.sum(tmp, 1).reshape(-1, 1), (1, tmp.shape[0]))      tmp = tmp / sumx      ret = np.inner(v1, np.dot(x, v2))      return ret

thanks!!

tl;dr: how-to linear algebra in cython?

the answer you link to still way call blas function cython. not python wrapper, python merely used c pointer function , can done @ initialization time. should c-like speed. wrong, think upcoming scipy 0.16 release provide convenient blas cython api, based on approach, not change things performance wise.

if didn't experience speed after porting cython of repeatedly called blas functions, either python overhead doing in numpy doesn't matter (e.g. if computation expensive part) or doing wrong (unnecessary memory copies etc.)

i approach should faster , easier maintain gsl, provided of course compiled scipy optimized blas (openblas, atlas, mkl, etc).


Comments

Post a Comment

Popular posts from this blog

qt - Using float or double for own QML classes -

-
i've created components , helper methods using c++ , qt used in qml gui. of methods calculate gui elements (size, transformations, scene graph items creation). target arm 32 bit using opengl / eglfs render output. opengl using float , qml using real (defined double ). what should use in c++ do not lose precision? if use qreal ( double ) shouldn't lose precision. if git grep float in qtbase.git , can find out bit more: dist/changes-5.2.0-**************************************************************************** dist/changes-5.2.0-* important behavior changes * dist/changes-5.2.0-**************************************************************************** dist/changes-5.2.0- dist/changes-5.2.0- - qt compiled qreal typedef'ed double on dist/changes-5.2.0: platforms. qreal float on arm chipsets before. guarantees more dist/changes-5.2.0- consistent behavior between platforms qt supports, binary dist/chan...
Read more

Create Outlook appointment via C# .Net -

-
i need create outlook appointment using microsoft.office.interop.outlook , while can work locally on own workstation , works if run web application via server's browser, not work when connect server externally. hence, think might permission issue. i changed application pool identity " localsystem " don't access denied error. unfortunately, doesn't work. the app behaves if appointment created, appointment doesn't appear in outlook. here include file on top of aspx.cs page using outlook = microsoft.office.interop.outlook; here code using pop appointment. outlook.application apptapp = new outlook.application(); outlook.appointmentitem appt = apptapp.createitem(outlook.olitemtype.olappointmentitem) outlook.appointmentitem; appt.subject = txtfirstname.text + " " + txtlastname.text; appt.body = txtcomment.text; appt.alldayevent = false; appt.start = datetime.parse(txtreminderdate.text + " 8:00 am"); appt.end = datetime.p...
Read more

ios - Swift Array Resetting Itself -

-
i have 2 arrays: pointhistory , datehistory . have nstimer , , when timer runs out, action called. in action appending int pointhistory , , current date string datehistory . here timer action: func timeraction() { let steppervalue = int(stepper.value) employee.numberofpoints = steppervalue networking.saveemployee(employee, completion: { (error) -> void in if error != nil { let alert = uialertcontroller(title: "error", message: "there error updating \(self.employee.name)'s points", preferredstyle: .alert) let alertaction = uialertaction(title: "ok", style: .cancel, handler: nil) alert.addaction(alertaction) self.presentviewcontroller(alert, animated: true, completion: nil) } }) var dateformatter = nsdateformatter() dateformatter.dateformat = "mm-dd-yyyy" dateformatter.timezone = nstimezone.localtimezone() var date = nsdate() var ...
Read more